36: Linear Regression | TensorFlow | Core API | Tutorial

The video discusses in TensorFlow core API: Linear regression Dataset: Auto MPG 00:00:00 - Overview 00:01:00 - Import libraries 00:02:40 - Download data: Auto MPG 00:04:22 - Download data: pd.read_csv() 00:05:27 - Data: check missing values: .isna().sum() 00:06:57 - Data: convert dataframe to tensor 00:08:24 - Data: Shuffle, split: tf.random.shuffle() 00:10:15 - Feature engineering: one-hot encoding 00:10:49 - Create function for one-hot encoding 00:14:58 - Data: One hot encode 00:17:09 - Normalize: create class Normalize(tf.Module) 00:19:54 - Normalize: x_train, y_train, x_test, y_test 00:21:42 - Build a machine learning model: create class LinearRegression(tf.Module) 00:25:24 - Check class LinearRegression() 00:26:24 - Predict (without training) 00:28:45 - Loss function: def mse_loss() 00:30:52 - Train and evaluate the model: set batch size 00:31:32 - Convert data: tf.data.Dataset.from_tensor_slices(), .shuffle(), .batch() 00:33:55 - Training loop: set parameters 00:35:29 - Training loop: begin 00:57:30 - Training loop: end 00:59:07 - Plot loss 01:01:22 - Saved model: create class ExportModule(tf.Module) 01:07:52 - Reload model: tf.saved_model() 01:10:09 - Ending notes # ---------------- # TensorFlow Guide # ---------------- https://www.tensorflow.org/guide/core/quickstart_core